RG11 - Rain rate calculation

[dive76]

Would it be complicated to do a rain rate calculation for the RG11 input? Would much welcome such an addition (webtag) to Cumulus new RG11 features.

Thanks,

Lars

[steve]

It would be like the Fine Offset calculation, based on the rain in the last five minutes. Quite a lot of extra code, and not very useful, in my opinion, like the FO figure. It's not really a very good way to calculate rain rate.

[dive76]

Ok, thanks for your comment. What was the experience you made with it for the FO stations. Why is this not a good way to calculate rain rate? Trying to learn a bit more....

[steve]

Why is this not a good way to calculate rain rate?

It needs to be 'dynamic' and based on the time between tips, like the way Davis weather stations do it. Unless the rain is heavy, the five-minute calculation gives sudden changes in rate, between a positive value and zero.

[dive76]

Thanks Steve, I get the point. I guess to implement such a dynamic rate calculation would be a lot of work to do.

[steve]

I guess to implement such a dynamic rate calculation would be a lot of work to do.

I did the write the code (for Fine Offset stations) and it's still there, dormant, in Cumulus. But there were some issues with it so I had to disable it. One day I'll look at it again and see if I can fix the problems.

[dive76]

Sorry Steve that I come back on this topic,
If I take it in a mathematical way the rain rate should be the incremental gradient of the rain time-volume curve.
If we assume that v1 would be the total amount of rain at a time t1 and v2 the total amount of rain at the next dip at time t2 then:
Rain rate=(v2-v1)/(t2-t1).

Or in more general terms for each step (dip) n:

Rain rate=(Vn+1 – Vn)/ (tn+1 –tn)

This would be a very simple and short algorithm to make.

I guess the topic is much for complex and I am most likely missing some important point here. Just try to learn a bit more

[steve]

It's not the calculation itself that is the issue, as I said, I have written code for it which works perfectly well. In general (ignoring the specific situation of the RG11) the hard part is catering for all the various combinations of live data and logger data, getting garbage from the station, starting and stopping Cumulus etc etc.

I'm not stupid, honestly; I just need to find time to look at it again and work out what the problem was.